Predicting Dickens: 7-12


Art piece (Carrie Roy): Sculpture, black walnut and printed paper 50” x 20” x 26”

A steamer trunk of data to explore the missing chapters of Dicken's The Mystery of Edwin Drood

Literary Interpretation (Catherine DeRose): Edwin Drood is famous for being Dickens's final, incomplete novel. Only 6 of the planned 12 installments were published. Since the release of the last installment, readers have produced numerous continuations that aim to solve the mystery. For Predicting Dickens, I calculated the likelihood for certain endings given patterns from Dickens's preceding novels. To calculate the likelihoods, I generated probabilities using all of Dickens's novels, recording things like: how many characters are guilty of murdering their relatives (a major plot point in Edwin Drood)? how many characters are innocent of murder? I then used this data to calculate out the likely fates of certain characters. In the next stage of this project, I will compare these statistical results against readers' intuitions and desires by looking at adaptions and continuations of the novel. For more information on the project, please send me an email. Spoiler alert, there is an 83% probability that Edwin Drood is alive!

Statistical Interpretation and Code: In the early stages of the project, we used a tool, Narrative Threads, that Carrie Roy helped develop, for added insight into the progression of Dickens's narratives. We looked at things like: average word word length, chapter length, installment length, character numbers, character gender, unique words, total words, narrative profile, and more.

Link to Dickens Narrative Visualizations Key
Link to Dickens Narrative Visualizations (created in Narrative Threads tool developed by Kevin Ponto, Carrie Roy, Eric Hoyt UW Madison)